Skip to content
AI.info

Model Tracker

Code Arena (WebDev) leaderboard

The models in the Model Tracker that Arena has measured on Code Arena (WebDev), ranked by the highest score of each: the name Arena gives it, the setting it ran and the day. Arena’s leaderboard dataset (lmarena-ai/leaderboard-dataset on Hug

The models in the Model Tracker that Arena has measured on Code Arena (WebDev), ranked by the highest score of each: the name Arena gives it, the setting it ran and the day. Arena’s leaderboard dataset (lmarena-ai/leaderboard-dataset on Hugging Face), CC BY 4.0: ratings from people’s votes between two anonymous models, as Arena publishes them.

  1. 1. Claude Opus 5.5: 1,806 (1,792–1,820) — Claude Opus 5.5 (Max), Max, 18 October 2026
  2. 2. GPT-6 Astra: 1,776 (1,765–1,786) — GPT-6 Astra (Max), Max, 18 October 2026
  3. 3. Claude Sonnet 5.5: 1,768 (1,755–1,781) — Claude Sonnet 5.5 (xHigh), xHigh, 18 October 2026
  4. 4. GPT-6.1 Sol: 1,754 (1,741–1,767) — GPT-6.1 Sol (Max), Max, 18 October 2026
  5. 5. Claude Fable 5.1: 1,742 (1,733–1,751) — Claude Fable 5.1 (Max), Max, 18 October 2026
  6. 6. Claude Opus 5: 1,690 (1,683–1,696) — Claude Opus 5 (Max), Max, 18 October 2026
  7. 7. GPT-6 Sol: 1,682 (1,672–1,693) — GPT-6 Sol (Max), Max, 18 October 2026
  8. 8. Gemini 4 Argon: 1,678 (1,663–1,694) — Gemini 4 Argon (High), High, 18 October 2026
  9. 9. Qwen3.8-Max: 1,673 (1,661–1,685) — Qwen3.8-Max, 18 October 2026
  10. 10. Muse Wonder: 1,650 (1,636–1,664) — Muse Wonder (Max), Max, 18 October 2026
  11. 11. Grok 4.7: 1,638 (1,629–1,648) — Grok 4.7 (xHigh), xHigh, 18 October 2026
  12. 12. Qwen3.8-Flash-Next: 1,629 (1,621–1,637) — Qwen3.8-Flash-Next, 18 October 2026
  13. 13. Claude Fable 5: 1,624 (1,618–1,631) — Claude Fable 5 (High), High, 18 October 2026
  14. 14. Grok 4.6: 1,617 (1,610–1,624) — Grok 4.6 (High), High, 18 October 2026
  15. 15. Qwen3.8-27B: 1,592 (1,585–1,599) — Qwen3.8-27B, 18 October 2026
  16. 16. Gemini 3.7 Flash: 1,590 (1,583–1,597) — Gemini 3.7 Flash (High), High, 18 October 2026
  17. 17. Claude Haiku 5.5: 1,586 (1,575–1,597) — Claude Haiku 5.5 (High), High, 18 October 2026
  18. 18. Gemini 3.8 Flash: 1,586 (1,578–1,593) — Gemini 3.8 Flash (High), High, 18 October 2026
  19. 19. DeepSeek-V4-Flash: 1,582 (1,573–1,592) — DeepSeek-V4-Flash (High), High, 18 October 2026
  20. 20. GPT-6 Luna: 1,577 (1,570–1,583) — GPT-6 Luna (Max), Max, 18 October 2026
  21. 21. Claude Opus 4.7: 1,558 (1,552–1,564) — Claude Opus 4.7 (High), High, 18 October 2026
  22. 22. Claude Opus 4.8: 1,556 (1,550–1,562) — Claude Opus 4.8 (High), High, 18 October 2026
  23. 23. Grok 4.5: 1,553 (1,546–1,560) — Grok 4.5, 18 October 2026
  24. 24. Claude Opus 4.6: 1,547 (1,541–1,552) — Claude Opus 4.6 (High), High, 18 October 2026
  25. 25. Mistral Large 4: 1,542 (1,530–1,555) — Mistral Large 4, 18 October 2026
  26. 26. Claude Sonnet 5: 1,541 (1,535–1,547) — Claude Sonnet 5 (High), High, 18 October 2026
  27. 27. Gemini 3.6 Flash: 1,538 (1,531–1,546) — Gemini 3.6 Flash (High), High, 18 October 2026
  28. 28. Claude Sonnet 4.6: 1,523 (1,518–1,528) — Claude Sonnet 4.6, 18 October 2026
  29. 29. Gemini 3.5 Flash: 1,499 (1,492–1,507) — Gemini 3.5 Flash (High), High, 18 October 2026
  30. 30. DeepSeek V4 Pro: 1,447 (1,441–1,454) — DeepSeek-V4-Pro, 18 October 2026
  31. 31. Gemini 3 Pro: 1,440 (1,432–1,449) — Gemini 3 Pro, 18 October 2026
  32. 32. Gemini 3 Flash: 1,440 (1,432–1,449) — Gemini 3 Flash, 18 October 2026
  33. 33. Gemini 3.5 Flash-Lite: 1,435 (1,394–1,476) — Gemini 3.5 Flash-Lite, 18 October 2026
  34. 34. GPT-5: 1,418 (1,402–1,433) — GPT-5 (Medium), Medium, 18 October 2026
  35. 35. GPT-5.2: 1,417 (1,394–1,439) — GPT-5.2, 18 October 2026
  36. 36. GPT-5.4: 1,401 (1,391–1,412) — GPT-5.4, 18 October 2026
  37. 37. GPT-5.4 Mini: 1,398 (1,391–1,404) — GPT-5.4 mini (High), High, 18 October 2026
  38. 38. GPT-5.1: 1,396 (1,384–1,407) — GPT-5.1 (Medium), Medium, 18 October 2026
  39. 39. Qwen3.5-122B-A10B: 1,361 (1,353–1,368) — Qwen3.5-122B-A10B, 18 October 2026
  40. 40. Qwen3.5-27B: 1,358 (1,350–1,366) — Qwen3.5-27B, 18 October 2026
  41. 41. Grok 4.3: 1,358 (1,352–1,364) — Grok 4.3, 18 October 2026
  42. 42. GPT-5.2-Codex: 1,339 (1,330–1,348) — GPT-5.2-Codex, 18 October 2026
  43. 43. GPT-5.1-Codex: 1,337 (1,325–1,349) — GPT-5.1-Codex, 18 October 2026
  44. 44. DeepSeek-V3.2-Exp: 1,272 (1,258–1,285) — DeepSeek-V3.2-Exp, 18 October 2026
  45. 45. Mistral Medium 3.5: 1,264 (1,250–1,279) — Mistral Medium 3.5, 18 October 2026
  46. 46. Qwen3.5-35B-A3B: 1,254 (1,237–1,272) — Qwen3.5-35B-A3B, 18 October 2026
  47. 47. GPT-5.1-Codex Mini: 1,244 (1,222–1,266) — GPT-5.1-Codex-Mini, 18 October 2026
  48. 48. Grok 4.1 Fast: 1,242 (1,231–1,253) — Grok 4.1 Fast (Reasoning), Reasoning, 18 October 2026
  49. 49. Mistral Large 3: 1,230 (1,205–1,256) — Mistral Large 3, 18 October 2026
  50. 50. Gemini 2.5 Pro: 1,226 (1,211–1,242) — Gemini 2.5 Pro, 18 October 2026
  51. 51. Grok 4.1: 1,215 (1,190–1,240) — Grok 4.1 (Thinking), Thinking, 18 October 2026
  52. 52. Devstral 2: 1,197 (1,176–1,217) — Devstral 2, 18 October 2026
  53. 53. Grok Code Fast 1: 1,166 (1,138–1,195) — Grok Code Fast 1, 18 October 2026
  54. 54. Grok 4 Fast: 1,160 (1,132–1,187) — Grok 4 Fast (Reasoning), Reasoning, 18 October 2026

Arena: the benchmark and its data · CC BY 4.0

Other benchmarks

All of the Model Tracker